Population genetics research has been conducted on the ancestry of the modern Turkish people (not to be confused with Turkic peoples) in Turkey. Such studies are relevant for the demographic history of the population as well as health reasons, such as for studying population-specific diseases. Some studies have sought to determine the relative genetic contributions of the Turkic peoples of Central Asia, from where the Seljuk Turks began migrating to Anatolia after the Battle of Manzikert in 1071, which led to the establishment of the Anatolian Seljuk Sultanate in the late 11th century, and prior populations in the area who were culturally assimilated during the Seljuk and the Ottoman periods.
Turkish genomic variation, along with several other West Asian populations, looks most similar to genomic variation of Southern European populations such as southern Italians. West Asian genomes, including Turkish ones, have been greatly influenced by early agricultural populations in the area; later population movements, such as those of Turkic speakers, also contributed. However, the genetic variation of various populations in Central Asia "has been poorly characterized"; Western Asian populations may also be "closely related to populations in the east".
Multiple studies have found similarities or common ancestry between Turkish people and present-day or historic populations in the Mediterranean, West Asia, and the Caucasus. Several studies have also found some Central Asian contributions.
Central Asian geneflow
Several studies have investigated to what extent a gene flow from Central Asia to Anatolia contributed to the gene pool of the Turkish people and the role of the 11th-century settlement by Oghuz Turks. Central Asia is home to numerous populations that "demonstrate an array of mixed anthropological features of East Eurasians (EEA) and West Eurasians (WEA)"; two studies showed Uyghurs have 40-53% ancestry classified as East Asian, with the rest being classified as European; another study put the European-related ancestry at 36%. A 2018 autosomal single-nucleotide polymorphism study suggested that Eurasian Steppe slowly transitioned from Indo-European and Iranian-speaking groups with largely western Eurasian ancestry to increasing East Asian ancestry with Turkic and Mongolic groups in the past 4,000 years, including extensive Turkic migrations out of Mongolia and slow assimilation of local populations.
Two earlier (2000 and 2002) studies suggested that, although the Turks' settlement of Anatolia was of cultural importance, including the introduction of the Turkish language and Islam, the genetic contribution from Central Asia may have been slight. A 2020 global study looking at whole-genome sequences showed that Turks have relatively lower within-population shared identical-by-descent genomic fragments compared to the rest of the world, suggesting mixture of remote populations.
A 2003 study found that some Xiongnu remains from Mongolia had paternal and maternal genetic lineages that have also been found in people from modern-day Turkey. Most (89%) of the Xiongnu sequences in this study belonged to Asian maternal haplogroups, however other studies have shown a significantly higher frequency of West Eurasian maternal and paternal haplogroups in Xiongnu samples, indicating a diverse population. Contact between West and East Eurasian populations pre-dates the Xiongnu period.
A study published in 2003 looked at human leukocyte antigen genes to investigate the affinity of certain Mongolian tribes with Germans and Anatolian Turks. It was found that Germans and Anatolian Turks were equally distant to the Mongolian populations. No close relationship was found between Anatolian Turks and Mongolians despite the history of contact and mutual influence between Turkic and Mongolic peoples.
A study of 75 individuals from various parts of Turkey concluded that the "genetic structure of the mitochondrial DNAs in the Turkish population bears some similarities to Turkic Central Asian populations".
A 2001 study comparing the populations of Mediterranean Europe and Turkic-speaking peoples of Central Asia estimated the Central Asian genetic contribution to current Anatolian Y-chromosome loci (one binary and six short tandem repeat) and mitochondrial DNA gene pool to be roughly 30%. A 2004 high-resolution SNP analysis of Y-chromosomal DNA in samples collected from blood banks, sperm banks, and university students in eight regions of Turkey found evidence for a weak but detectable signal (<9%) of recent paternal gene flow from Central Asia. However, this received serious criticism for not including non-East Eurasian Central Asian lineages, such as specific R1 and J clades, which are commonly found in Western Turkic populations.Y-chromosome distributions among populations in Northwest China identify significant contribution from Central Asian pastoralists and lesser influence of western Eurasians A 2006 study concluded that the true Central Asian contributions to Anatolia was 13% for males and 22% for females (with wide ranges of confidence intervals), and the language replacement in Turkey and might not have been in accordance with the elite dominance model. It was later observed that the male contribution from Central Asia to the Turkish population with reference to the Balkans was 13%. For all non-Turkic speaking populations, the Central Asian contribution was higher than in Turkey. According to the study, "the contributions ranging between 13%–58% must be considered with a caution because they harbor uncertainties about the state of pre-nomadic invasion and further local movements."
In a 2015 study, Turkish samples were in the West Eurasian clade which consisted of "all of mainland Europe, Sardinia, Sicily, Cyprus, western Russia, the Caucasus, Turkey, and Iran, and some individuals from Tajikistan and Turkmenistan." In this study, ancestry from East Asia was also visible in Turkish samples, with events after 1000 CE generally involving Asian sources being important when it comes to the ancestry of Turkey and its region.
As of 2017, Central Asian genetic variation has been poorly studied, with little or no whole genome sequencing data for countries such as Turkmenistan and Afghanistan. Therefore, future comprehensive genome-wide studies are needed. Turkish people, and other West Asian populations, may also be closely related to Central Asian populations such as those near West Asia.
A 2022 archaeogenetic study by Lazaridis et al., analyzing 58 individuals originally sourced from a 1995 study across six Turkish cities, used two-way ADMIXTURE analysis of East Eurasian ancestry to track Central Asian input into Anatolia. This ancestry was absent (0%) in Byzantine-era Anatolian individuals, but present at approximately 9% in present-day Turks and approximately 18% in a transitional population of the 14th–17th centuries from the Aegean coastal site of Çapalıbağ. Modelling present-day Turks and the Çapalıbağ population as mixtures of Roman/Byzantine-era Anatolians and Central Asians, the authors estimated that admixture occurred about 12 generations (roughly 340 years) before the Çapalıbağ individuals' time and about 30 generations (roughly 860 years) ago for present-day Turks—dates they linked to the arrival and expansion of the Seljuk Turks in Anatolia from the late 11th century onward. In this model the local, non–Central Asian component of present-day Turkish ancestry was represented by Roman/Byzantine-era Anatolians, and the authors placed the overall Central Asian Turkic contribution at roughly 9–22%.
Haplogroup distributions
A 2021 study which looked at whole genomes and whole-exomes of 3,362 Turkish people found that the most common Y chromosome haplogroups were J2a, R1b, and R1a (18.4%, 14.9%, and 12.1% respectively). Haplogroups C-M130 and O3 ranged from 8.5% to 15.6%. Most common mtDNA haplogroups were H, U, and T (27.55%, 19.53%, and 10.99% respectively).
An earlier 2004 study of 523 people found many Y-DNA haplogroups in Turkey. Most haplogroups in Turkey are shared with its West Asian and Caucasian neighbors. The most common haplogroup in Turkey is J2 (24%), which is widespread among Mediterranean, Caucasian, and West Asian populations. Haplogroups that are common in Europe (R1b and I; 20%), South Asia (L, R2, H; 5.7%), and Africa (A, E3*, E3a; 1%) are also present. By contrast, Central Asian haplogroups (C, Q, and O) are rarer. While, K, R1a, R1b, and L infrequently occur in Central Asia, they are notable in many other Western Turkic groups. J2 is also found in Central Asia, a notably high frequency (30.4%) being observed among Uzbeks.
However, because the dataset relies on low-resolution macro-haplogroups, attempting to quantify paternal flow from Turkic-related Central Asian migrations yields a highly ambiguous range. By excluding branches that are very unlikely to stem from these migrations and including those that potentially do, the contribution spans from 8.6% to 55.5%. The 8.6% minimum includes definitively East Eurasian and Central Asian clades (C*-M216, C3-M217, N*-M231, N3a-M178, O3-M122, Q*-M242, Q2-M25, R1b2-M73). The 55.5% maximum also includes clades with less certain origins, including some J, G, R1a, and R1b markers, which occur in Central Asia but are also found among indigenous West Asian populations. Thus, this paper is of limited use for determining the actual extent of Turkic migration, since the broad macro-haplogroup categories cannot reliably distinguish Central Asian ancestry introduced by Turkic migrations from lineages already present in West Asia.
The main percentages of Y chromosome haplogroups identified in the 2004 study were as follows:
J2: 24%. J2 (M172) may reflect the spread of Anatolian farmers. J2-M172 is "mainly confined to the Mediterranean coastal areas, southeastern Europe and Anatolia", as well as West Asia and Central Asia. R1b: 15.9%. R1b is found mainly in Europe, also West Asia, Central Asia, Southern Asia, some parts of the Sahel region of Africa. G: 10.9%. Haplogroup G has also been associated with the spread of agriculture (together with J2 clades) and is "largely restricted to populations of the Caucasus and the Near/Middle East and southern Europe." The G2a subclade in particular is associated with the Early European Farmers, who in turn descend from the Anatolian farmers. E3b-M35: 10.7% (E3b1-M78 and E3b3-M123 accounting for all E representatives in the sample, besides a single E3b2-M81 chromosome). E-M78 is common along a line from the Horn of Africa via Egypt to the Balkans. Haplogroup E-M123 is found in both Africa and Eurasia. J1: 9% R1a: 6.9% I: 5.3% K: 4.5% L: 4.2% N: 3.8% T: 2.5% Q: 1.9% C: 1.3% R2: 0.96%
Other markers that occurred in less than 1% are H, A, E3a, O, and R1*.
A 2011 study took into account oral histories and historical records. The researchers went to four settlements in Central Anatolia and chose a random selection of subjects from among university students.
In an Afshar village near Ankara where, according to oral tradition, the ancestors of the inhabitants came from Central Asia, the researchers found that 57% of the villagers had haplogroup L, 13% had haplogroup Q and 3% had haplogroup N. The high rate of haplogroup L observed in this study, which is most common in South Asia, was difficult for researchers to explain and could not be traced back to any specific geographic location, and authors said it would be difficult to associate this haplogroup with the Turkic migrations, given the paucity of evidence. Furthermore, 10% of the Afshars had haplogroups E3a and E3b, while only 13% had haplogroup J2a, the most common in Turkey.
By contrast, the inhabitants of a traditional Turkish village that had little migration had about 25% haplogroup N and 25% J2a, with 3% G and close to 30% R1 variants (mostly R1b).
Whole genome sequencing